Upper mantle anisotropy of the Borborema Province, NE Brazil: Implications for intra-plate deformation and sub-cratonic asthenospheric flow

Abstract

The geological record of the Borborema Province, northeast Brazil, documents tectonic events that characterised the Precambrian formation and Mesozoic breakup of Gondwana. Large-scale shear zones and associated granitic plutons that developed during the Neoproterozoic Brasiliano/Pan-African orogeny, and major sedimentary basins of Mesozoic age, indicate significant deformation across the region. However, whether or not the shear zones resulted from Precambrian terrane accretion, or are simply the result of episodes of subsequent intra-plate deformation is debated. Also poorly understood is the effect of the thick São Francisco mantle keel on present-day asthenospheric flow. To address these issues we have performed a teleseismic shear wave splitting study of mantle seismic anisotropy from a new broadband seismograph network centred on the Borborema Province. Shear wave splitting parameters (φ, δt) reveal a lack of plate-scale anisotropic fabrics associated within the continental interior, perhaps supporting models of formation and evolution of the Borborema Province involving minimal deformation of the lithospheric mantle. Delamination of anisotropic lithosphere during the development of Cenozoic volcanism that eroded older fossil lithospheric fabrics is unlikely because the widespread Cenozoic magmatism required to achieve this is absent in the geological record. Instead, the apparently low levels of seismic anisotropy observed in the interior of the Borborema Province may simply reflect depth-dependent anisotropy: nulls/low δt observations may be the subtractive result of orthogonal fast directions in the lithosphere and asthenosphere. Towards the Brazilian coast δt>1s, and fast directions are sub-parallel to stretching fabrics formed during the opening of the South Atlantic. This may imply that the mantle lithosphere was deformed but not completely destroyed during Gondwana breakup. However, a more complete backazimuthal coverage of splitting measurements across the region is needed to confirm or refute these hypotheses unambiguously.